This is an open access article distributed under the terms of the Creative Commons
Attribution License (In Europe, as well in most of the rest of world, the outbreak of the COVID-19 pandemic led several countries to adopt measures restricting individual mobility and, above all, entailing a ban on private and public gatherings. Violations of such restrictions (such as illegal gatherings) were relatively rare, though some episodes hit the headlines, especially involving young people who gathered mainly for the purpose of having fun, drinking, and dancing with friends. Such events were often treated by the law enforcers as threats to public safety and police intervention was deemed necessary to disperse the participants. For instance, on 1 April 2021 in Brussels (Belgium), police officers dispersed nearly 2,000 people using water cannons, a measure considered too heavy-handed and inappropriate in the case of peaceful and non-violent gatherings (OSCE Office for Democratic Institutions and Human Rights, 2019). As Human Rights Watch has noted, the 2020 United Nations guidelines on the use of less-lethal weapons states that the “water cannon should only be used in situations of serious public disorder where there is a significant likelihood of loss of life, serious injury, or the widespread destruction of property” (United Nations Human Rights, 2020, p. 38). “They should only be used if violence has reached such a level that targeting individuals engaged in violence is not a possible or sufficient response” (OSCE Office for Democratic Institutions and Human Rights, 2019, p. 71). Moreover, their use to disperse crowds has been criticized as it can increase panic resulting in further injuries and an escalation of violence (OSCE Office for Democratic Institutions and Human Rights, 2019; World Organisation Against Torture, 2021). All these organizations have suggested that their use is at least inappropriate to disperse peaceful and nonviolent gatherings.
The aim of this research was to investigate whether people in times of pandemic were likely to consider as acceptable controversial measures imposed to disperse peaceful but illegal gatherings. In line with previous research (Moghaddam, 2016; Passini, 2017b; Passini & Morselli, 2010), in this study we investigated whether the acceptance of such measures might be intertwined with the perceived legitimacy of the political system in which the event occurred. In particular, we inquired whether controversial measures might be paradoxically considered more acceptable if implemented by democratic states, as compared with authoritarian or autocratic countries. This democratic delusion paradox1 has been explained as the tendency to assume that a state that claims to be democratic will never adopt measures that are undemocratic or inappropriate in terms of their severity (Kelman & Hamilton, 1989; Moghaddam, 2016; Passini & Morselli, 2013). In line with Kelman and Hamilton’s (1989) political orientation theory, it was hypothesized that people without a value-based orientation to authority should be those who may fall into this paradox. Instead, value-oriented citizens should be able to assess the severity of the measure taken by the authority, regardless of the perceived level of democracy of the authority itself. Indeed, the novelty of the present research was to analyze obedience to policies demanded by an authority using a theory that considers not only the perceived legitimacy of authority, but also the specific legitimacy of the demands themselves.
The research took place in Italy, one of the European countries most affected by the pandemic and that had introduced among the tightest restrictions to combat it (Hale et al., 2020). Though at the beginning, the legitimacy of these anti-COVID measures were debated within the Catholic church (see Colaianni, 2020) and by some Italian scholars (e.g. Licastro, 2020), the restrictions were accepted without prompting large scale protests, as they were considered useful to offset the disease spread (Carlucci et al., 2020; Turina, 2021). Violations of anti-COVID measures were indeed harshly judged by both the media and the people, with several cases of policing by private citizens (Scalia, 2021).2
Concern for COVID-19 and Support for Controversial Measures
Since the beginning of COVID-19 and the increase in documented casualties, governments have adopted a series of measures to counter the pandemic, particularly by restricting gatherings and citizens’ mobility (also referred to as lockdown). While anti-COVID restrictions were effective in Italy as well in other parts of the world to reduce contagions and casualties (Pelagatti & Maranzano, 2021), some organizations have raised ethical problems when excessive force was used to resolve a health issue and fundamental democratic rights, such as freedom of expression and right to dissent, were jeopardized (Altshuler, 2021; Human Rights Watch, 2021; Office of the High Commissioner for Human Rights, 2020; World Organisation Against Torture, 2021). Amnesty International (2020) has raised concerns about the excessive use of force by the police in Europe, such as cases of beating of citizens violating the mobility restrictions in Italy. Though policing was necessary to ensure compliance with the lockdown measures to combat the pandemic, “the use of force against individuals who did not pose any threat to law enforcement officials appears to be disproportionate to achieve the legitimate aim of protecting public health” (Amnesty International, 2020, p. 24).
The so-called democratic delusion paradox, already put forward by the analysis of Cheibub and colleagues (2020), is that citizens in democracies might condemn the same harsh actions if they take place in countries perceived as dictatorial, while they would support – or at least not oppose – them if they take place in contexts perceived as democratic. As some researchers (Moghaddam, 2016; Passini, 2017b, 2022; Passini & Morselli, 2010, 2011) have pointed out, in liberal democracies people often have an overestimated perception of their liberties and at times do not recognize their possible restrictions. Nevertheless, also in Western democratic countries, civil liberties have been restricted, especially in reaction to threats, ringing some alarm bells by human-rights organizations and scholars (Fitzpatrick, 2003; Grayling, 2010). For instance, some acts have been introduced in many countries after the 9/11 attacks to detain foreign citizens without formal charges. Without entering the debate about the rightfulness and utility of such measures, it is worth focusing on those variables that may affect citizens’ obedience to these. As the long tradition of studies on blind obedience to authority has shown (see Blass, 1999, 2001; Moghaddam, 2016), even societies based on a so-called democratic system can degenerate into autocracies where there is no active participation and critical control by the citizens in the policies being enforced.
In analyzing what might lead people to comply with and support requests from authority without any criticism or questioning, the psychosocial literature has mainly focused on political trust. Various studies (e.g. Devine et al., 2021; Dohle et al., 2020; Jäckle et al., 2022) have indeed considered how political trust is a variable that can explain how people accept authority’s demands and obey laws without questioning their validity and legitimacy. These studies have analyzed the impact of political trust also in the case of anti-pandemic policies. Jäckle and colleagues (2022) have underlined that under conditions of uncertainty political trust can serve as a decision-making heuristic in following or not following the directives and demands of institutions. However, in this process it is the source of the request and not the request itself that is judged. Therefore, political trust can lead to suspend judgement on the policies themselves. Johansson and colleagues (2021) have argued that this was the case for most of the Swedish population following the government’s choice not to mandate the use of face masks during the COVID-19 pandemic, with the ensuing increase in infections and deaths. The issue is that by placing a high level of trust in the government or authority in office, individuals would indeed be unable to recognize whether the demands coming from the authority itself conflict with the common good and the democratic values on which democracy should be based. That is, trusting citizens may take for granted that the authority upholds democratic values and may thus fall into the previously mentioned democratic delusion paradox.
In order to better understand the risks of degeneration of democratic systems and how to prevent them from occurring, a theory that also considers the legitimacy of authority’s demands is needed. In this regard, the political orientation theory of Kelman and Hamilton (1989; Morselli & Passini, 2015) may be helpful, as it focuses on the evaluation of two legitimacies: the legitimacy of the source of the demand (e.g., the institutions or the authority may or may not be recognized as legitimate) and the legitimacy of the demand itself (e.g., in a democratic system, whether such demand pursues the values of the system itself). According to Kelman and Hamilton, most people focus only on the legitimacy of the authority, without considering the legitimacy of the demand. The relevant point in political orientation theory is the consideration of individual differences in evaluating legitimacy. Some individuals may indeed evaluate the adequacy of authority’s policies regardless of whether it claims to be democratic or not. Specifically, the authors identified three distinct orientations (rule, role, and value) of individuals by which authority is experienced, assessed, supported, or opposed. These three orientations characterize different ways in which citizens relate to the political system, with implications for their own attribution of responsibility and participation in civil society. Although these orientations are not mutually exclusive – since they are likely to appear at some level in all individuals and they may coexist or alternate (Kelman & Hamilton, 1989) – they underline systematic differences in the way in which individuals define themselves as citizens and relate to the political system. Thus, the three orientations should be treated as ideal types with implications for the citizen-authority relationship (Kelman, 2006). Specifically, rule-oriented citizens tend to consider it their duty to comply with the authority’s rules; they expect the authority to protect their basic interests and ensure societal order, and their participation tends to be passive. Role-oriented citizens expect citizens to actively support the authorities in every situation; they expect the authorities to ensure their status and roles, and their participation tends to be supportive. Value-oriented citizens tend to be active in evaluating the authority’s policies; they expect the authority to pursue policies that reflect the fundamental values of a just and fair society (Morselli & Passini, 2015).
Thus, Kelman and Hamilton’s (1989) political orientation theory may be a useful tool to explain the response to policies. Indeed, while rule-, role-, and value-oriented citizens do not differ much in response to a legitimate authority’s demand as they will follow it, they do instead differ in the case of an illegitimate and controversial claim. Specifically, value-oriented citizens will be the most likely to recognize the illegitimacy of the demand, regardless of the legitimacy/illegitimacy of the authority itself, and to protest against this specific claim. This is because these individuals view the role of the citizen as encompassing the responsibility to evaluate and possibly question the authority’s demands. Instead, both rule- and role-oriented citizens will oppose policies only when they pose a threat to their security (rule-oriented) or status (role-oriented). For instance, migration policies might be perceived by role-oriented citizens as threatening the social fabric and enhance anomy, and by rule-oriented citizens as destabilizing social hierarchies. In either case, we could expect rule- and role-oriented citizens to oppose the political authority.
Hypotheses
The aim of the present research was to analyze, in times of the COVID-19 pandemic, people’s willingness to accept the implementation of controversial measures to stop the virus from spreading. Specifically, we expected (Hypothesis 1) a positive correlation between concerns about COVID-19 and the acceptance of the measures adopted. Indeed, concerns about the spread of the virus and fears for one’s safety should lead people to accept even more severe measures to limit the disease spread and punish offenders. Moreover, in accordance with political orientation theory, we expected (Hypothesis 2) people with a low value orientation to support more controversial measures to a greater extent in democratic contexts than in contexts perceived to be non-democratic, whereas people with a high value orientation will not make this differentiation and will not support severe measures in either context. The effect of value orientation will be controlled for ideological beliefs, such as authoritarianism and political affiliation. Past research (see Cohrs et al., 2005; Passini & Villano, 2018; Werts et al., 2013) has shown these are relevant factors influencing the acceptance of policies issued by the authority regardless of their legitimacy.
Method
Participants
The participants were contacted online, using an Internet questionnaire constructed using Limesurvey, a survey-generating tool (http://www.limesurvey.org). The participants were recruited by means of a snowballing procedure. Undergraduate students from the University of Bologna were asked to recruit adult individuals. No particular instructions were given to them, other than that the participants were of legal age and consenting. Respondents were informed that their participation was voluntary and that their responses would remain anonymous and confidential. No fee was offered. The questionnaire was drafted in Italian. The subject’s IP address was monitored in order to make sure no one re-entered the survey site. The data were collected in May 2021.
A total of 359 Italian citizens (67.1% women) responded by accessing the website and filling out the questionnaire. Participant ages ranged from 18 to 85 years (M = 36.19, SD = 15.76). They were mainly born in the north of Italy (68.7%), while 26.8% and 3.4% came from the center and the south, respectively, while 1.1% were born abroad. As regards their level of education, 5.3% declared they had completed middle school, 57.3% declared they had earned a high school diploma, 31.7% a Bachelor’s degree and 5.7% a Master’s or Ph.D. qualification. Job-wise, 42.3% stated they were white collar workers, 28.7% university students, 11.4% self-employed, 10.2% student worker, 2% unemployed, and, finally, 5.4% pensioners.
Design and Procedure
Participants first responded to socio-demographic questions. Then they were asked to assess the level of democracy of five countries and they responded to some questions regarding concerns over COVID-19. Afterwards, they read a make-believe news story as if it had appeared on the website of ANSA (Agenzia Nazionale Stampa Associata), the leading multimedia news agency in Italy (see Appendix). The report stated that the police used water cannons to disperse a gathering of young people who were in the university area dancing and drinking, almost all without wearing a face mask and in defiance of the bans related to the COVID-19 pandemic. Depending on the experimental condition (to which participants were randomly assigned), the city where the news took place was either Stockholm (n = 186) or Moscow (n = 173), the former being a city in a country usually considered democratic, and the latter less so.3 On purpose, the two cities were chosen because they did not belong to the participants’ life context. Then, the participants responded to the political orientation scale and to an authoritarianism scale.
Measures
Demographics and Political Affiliation
Participants indicated their age, sex, place of birth (Northern, Central, or Southern Italy), level of education and job. Moreover, they indicated their political affiliation (from 1 = extreme left to 10 = extreme right).
Level of Democracy
Participants were asked to assess the level of democracy of five countries (namely France, Hungary, Italy, Russia, Sweden) on a seven-point scale (ranging from 1 = not at all to 7 = very much).
Acceptance of the Use of Water Cannons
After reading the make-believe news story regarding the police reaction to the illegal gathering, participants were asked on a seven-point scale (ranging from 1 = not at all to 7 = very much), “To what extent do you agree with the police using water cannons to disperse the crowd?”
Concerns for COVID-19
Participants were asked to respond on a seven-point scale (ranging from 1 = not at all to 7 = very much) to four questions about the COVID-19 pandemic: “How concerned are you about the COVID-19 pandemic?,” “How exposed do you feel about the risk of coronavirus infection?,” “How worried are you about your loved ones with respect to the risk of infection?,” “To what extent do you fear for your life because of a COVID-19 infection?” These items were combined to compute a COVID-19 concerns index (α = .81).
Political Orientation Scale
A 12-item version of the political orientation scale (Kelman & Hamilton, 1989) was used. The scale measures the three orientations each with four items on a seven-point scale (ranging from 1 = not at all to 7 = very much): rule (e.g. “All one should expect from the government is that all citizens are able to earn a living and feel safe, ” α = .59); role (e.g. “The most valuable contribution each citizen can make is to actively support government policies, ” α = .76); and value orientation (e.g. “The most valuable contribution each citizen can make is to maintain an active approach and always question government policies, ” α = .60). This structure was upheld by a confirmatory factor analysis (CFA) computed with Mplus 7.1 (Muthen & Muthen, 2012). The three-factor model fitted the data well: χ2(48) = 111.99, CFI = 0.93, TLI = 0.90, RMSEA = .06, SRMR = .06 (see Supplementary Materials – Section 1 for all items and CFA results).
Authoritarianism
This construct was measured by an Italian 12-item scale based on Right-Wing Authoritarianism (RWA, Altemeyer, 1996). The scale was constructed and validated by Passini (2017a). The subjects responded to each item on a 7-point scale ranging from 1 (strongly disagree) to 7 (strongly agree). Sample items are “our country will be great if we do what the authorities tell us to do” and “our government has to eliminate any opposition” (see Supplementary Materials – Section 2 for all items). A total score was computed (α = .85).
Results
Participants considered Sweden (M = 5.48) to be more democratic than Russia (M = 2.11), t = 21.06, p < .001. The other countries obtained the following scores: France (M = 5.00), Hungary (M = 3.05), and Italy (M = 4.55). The level of acceptance of the use of water cannons in Stockholm and in Moscow (M = 4.13, M = 3.74, respectively) was not statistically different (F1, 358 = 3.51), although close to significance (p = .06).
As can be seen in Table 1, participants expressed medium to high concerns for COVID-19. On the political orientation scale, participants had high scores in value orientation, but lower scores on rule and role orientations. Finally, they had low scores on RWA attitudes and tended to be politically affiliated to the center-left. Bivariate correlations showed that acceptance of the police intervention was positively correlated to COVID-19 concerns (in line with Hypothesis 1), rule and role orientations, authoritarianism, and political affiliation, while negatively correlated to value orientation. Age was not significantly related with acceptance.
Table 1
Means, Standard Deviations and Pearson Correlation Coefficients (N = 359)
| Variable | M | SD | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
|---|---|---|---|---|---|---|---|---|---|
| 1. COVID-19 concerns | 4.40 | 1.25 | – | ||||||
| 2. Acceptance | 3.94 | 2.00 | .29*** | – | |||||
| 3. Rule orientation | 3.39 | 1.16 | .10 | .24*** | – | ||||
| 4. Role orientation | 2.88 | 1.18 | .16** | .22*** | .53*** | – | |||
| 5. Value orientation | 5.34 | 1.02 | .02 | -.23*** | -.09 | -.05 | – | ||
| 6. RWA | 2.99 | 1.06 | .13** | .34*** | .59*** | .52*** | -.22*** | – | |
| 7. Political affiliation | 4.34 | 1.83 | -.05 | .26*** | .33*** | .22*** | -.32*** | .59*** | – |
| 8. Age | 36.19 | 15.76 | .13** | .02 | .01 | .02 | .14** | .13** | .00 |
Note. Scores ranged from 1 to 7 except for political orientation (from 1 to 10) and age (from 18 to 85).
*p < .05. **p < .01. ***p < .001.
In order to test the hypotheses, a stepwise regression analysis with condition, COVID-19 concerns (Step 1) and their interaction (Step 2), the three political orientations (Step 3) and their interactions with the condition (Step 4), political affiliation and RWA (Step 5), and their interactions with the condition (Step 6) as independent variables and acceptance as dependent variable was computed. In line with Hypothesis 1, which predicted greater acceptance under threat to safety, Table 2 showed a positive relation between concerns over COVID-19 and acceptance of the measures in all models, indicating a higher acceptance when people worried more about the pandemic. This association was higher in the Swedish condition, compared to the Russian one (Step 2). Secondly, higher scores in value orientation were negatively associated with acceptance (Step 3), even when controlling for their perception of threat, political affiliation and authoritarian attitudes (Step 5). In line with Hypothesis 2, in Step 4 the condition × value orientation effect was significant. As concerns the control variables, in Step 6 three interaction effects were significant: condition × COVID-19, condition × value orientation, and condition × political affiliation.
Table 2
Linear Regression on Acceptance to the Police Intervention (N = 359)
| Predictor | b on Acceptance
|
|||||
|---|---|---|---|---|---|---|
| Step 1 | Step 2 | Step 3 | Step 4 | Step 5 | Step 6 | |
| Condition | -.10* | -.10* | -.10* | -.10* | -.08 | -.08 |
| COVID-19 concerns | .28*** | .28*** | .26*** | .26*** | .26*** | .25*** |
| Condition × COVID-19 | -.11* | -.08 | -.09 | -.11* | -.10* | |
| Rule orientation | .15** | .16** | .06 | .03 | ||
| Role orientation | .06 | .05 | .00 | -.03 | ||
| Value orientation | -.22*** | -.21*** | -.16** | -.15** | ||
| Condition × Rule orientation | .04 | .04 | -.02 | |||
| Condition × Role orientation | .03 | .01 | -.02 | |||
| Condition × Value orientation | .10* | .10* | .14** | |||
| Political affiliation | .08 | .10 | ||||
| Authoritarianism | .18* | .22** | ||||
| Condition × Political affiliation | .16** | |||||
| Condition × Author. | .04 | |||||
| R2 | .09 | .10 | .19 | .20 | .24 | .26 |
| F | 16.71*** | 12.77*** | 13.31*** | 9.53*** | 9.37*** | 9.01*** |
Note. Coefficients are standardized. Condition is Stockholm = 0, Moscow = 1.
*p < .05. **p < .01. ***p < .001.
Simple slope analyses were conducted to illustrate the nature of the interactions reported in Table 2. Figures 1, 2 and 3 provide the levels of acceptance in the two conditions at high and low ends (±1 SD) of COVID-19 concerns, value orientation, and political affiliation. The results showed that participants with high scores on COVID-19 concerns, a left-wing political affiliation, and low scores on value orientation accepted the police intervention more in the Stockholm condition than in the Moscow one (simple slopes: t = -2.76, p = .006; t = -2.85, p = .005; t = -2.96, p = .003, respectively). Conversely, participants with low scores on COVID-19 concerns, political preference to the right, and high scores on value orientation showed no significant difference in acceptance between the Stockholm and the Moscow news (simple slopes: t = 0.07, p = .94; t = 0.82, p = .41; t = -0.05, p = .96, respectively).
Figure 1
Levels of Acceptance in the Two Conditions at High and Low Ends (±1 SD) of COVID-19 Concerns
Figure 2
Levels of Acceptance in the Two Conditions at High and Low Ends (±1 SD) of Value Orientation
Figure 3
Levels of Acceptance in the Two Conditions at High and Low Ends (±1 SD) of Political Affiliation
Discussion
In this study, we investigated whether the democratic delusion paradox was supported in the context of a real crisis. Previously, researchers had argued from a theoretical viewpoint that people are more willing to accept undemocratic measures when they are endorsed by democratic authorities. Empirical support for this paradox had been found using hypothetical scenarios (Passini & Morselli, 2010, 2011), in which participants were to judge some vignette or some piece of news that did not directly affect their lives. The present research was instead embedded in a real-life context. Although the scenario that participants were asked to comment on was still about a relatively distant context to the participants (Sweden or Russia), it concerned a real-life situation that all participants had experienced personally. Italy was indeed one of the countries most affected by the first wave of the COVID-19 pandemic and had introduced some of the toughest measures in Europe (Hale et al., 2020).
In line with the literature on the connection between threat perception and authoritarianism (Duckitt, 2013), our findings show that higher concerns for COVID-19 are associated with a higher acceptance of the appropriateness of the use of water cannons, independent of the perception of the level of democracy in the context. Moreover, our findings on the interaction between concerns about COVID-19 and the experimental condition also corroborate the democratic delusion paradox showing that, though higher concerns were always associated with higher acceptance, the difference in acceptance was smaller when the evaluation took place in an authoritarian context.
The democratic delusion paradox can be explained by several factors that played a role in shifting the focus of the judgment from the content of the event (the controversiality of the measure) to its perpetrator (the country affiliation). First, it is likely that preconceptions about either country were at play. Although a favorable perception of Russia has increased in Italy over the past decade,4 especially among the right-wing, most Italians still see Russia in a negative light (Huang, 2020). Sweden has instead a history of being considered a highly advanced country, particularly from a social welfare and democratic values perspective (Borioni, 2017). Therefore, negative stereotypes, on the one side, and positive ones on the other, might have played a role in weakening the judgment on the harshness of the measures.
Furthermore, the results of this study show possible inoculations against the democratic delusion paradox. The simple slope analysis displayed in Figure 2 shows that value-oriented citizens were less influenced by the condition, even after controlling for their worries about COVID-19 and their political ideology. In other words, they were less focused on the legitimacy of the source (i.e. the authority) and more on the act itself (i.e. the policy). This result shows that respondents upholding humanitarian values and having an active and participatory political orientation are more focused on judging the behavior of the authority, notwithstanding who or what the authority represents. Thus, our results corroborate Kelman and Hamilton’s (1989) argument that value-oriented citizens could help societies to avoid falling into “blind” obedience and to keep a critical eye on the actions of the authority. That is, they go beyond a blind trust in the belief that every measure and policy issued by a democratic authority is necessarily good or cannot be challenged. Schatz and colleagues (1999) have also defined this kind of political participation as constructive (in opposition to blind) patriotism highlighting that being watchful vis-à-vis the authority does not necessarily imply to be anti-authority. Whereas constructive patriotism is based on judgment of the authority demands, blind patriots tend to accept them notwithstanding their content.
In this dynamic, the present results also suggest the role of the “democracy” label. Being socially accepted, “democracy” can be used as a mean to justify different types of policies, even those that are overtly unaligned with the democratic values and principles. This aspect resonates with system justification theory (Jost, 2019; Jost & Banaji, 1994), according to which positive qualities are attached to sociopolitical systems to legitimize aspects of existing social, economic, and political systems. Possibly in contrast with this theory, we argue that the use of labels as democracy are not overtly used by respondents to build and defend a “false consciousness” (Jost, 2019). Instead, they induce the delusion of rightfulness and legitimacy of the authority, facilitating compliance.
As concerns control variables, both authoritarian and right-wing oriented respondents accepted the police intervention in both conditions to a higher degree. Since, however, the interaction between authoritarianism and condition was not significant (indicating a linear relationship), it is interesting to interpret the interaction with political affiliation. Indeed, in the democratic condition (Stockholm) there were no differences in acceptance between left and right-wing supporters. However, in the less democratic condition (Russia) the tough measures were opposed by respondents with a left-wing leaning and accepted by those with a right-wing one. The democratic delusion paradox is thus more an issue for left-wing respondents, as right-wing individuals accept repressive actions to a greater extent in both cases (akin to the results concerning authoritarianism).
This study had some limitations which need to be addressed. Firstly, the results are limited with respect to the convenience sample and the socio-cultural context where the survey was administered (i.e. Italy). Further studies are needed to test, and eventually generalize, the democratic delusion paradox in other countries. This also raises questions about the generalizability of the results, which cannot be tested given the changing pandemic situation. We believe that, despite some differences in the levels of causalities and policies, the effect of the pandemic in Italy shared similarities with other countries. Most of the countries around the world did indeed introduce restrictions on individual freedoms, such as mobility and social contact (e.g. Moodley et al., 2020; Zolka et al., 2021). Though necessary, these restrictions could turn into abuses if enforced unconditionally and for extended periods. Secondly, we might wonder what would have changed if Italians had evaluated Italy as the country in which the event occurred. We deliberately decided not to ask to judge one’s own country as this evaluation could more easily be affected by the political affiliation of the participants and their assessment of the government in office at the time. That said, participants of the present research gave an average value as concerns Italy’s level of democracy, although the average was closer to that of Sweden than Russia. We might thus expect an acceptance of the use of water cannons more similar to Sweden, although other more specific factors may come into play in the evaluation of one’s own country depending, for instance, on political preferences or the performance of the government in charge. Thirdly, even tougher measures of police intervention could be investigated in the future to see if differences in acceptance between democratic and non-democratic contexts decline or increase. Fourthly, it would have been appropriate to study the reaction to the news set in other countries with different scores on the democracy index than Sweden and Russia. Unfortunately, the changing pandemic situation, currently receding fortunately, makes it inappropriate to replicate the exact conditions analyzed at the time of the data collection. However, our results are in line with previous research (Passini & Morselli, 2010, 2011) providing some external validity to the democratic delusion paradox. The added value of the present research is to embed these results in a real-world crisis, which has drastically affected everyone’s lives.
Notwithstanding these limitations, our study shows that leveraging on a realistically threatening context, people tend to overlook the actions of the authorities, when the authorities are considered as legitimate. This legitimacy can be built, among other factors, on positive prejudice but also on values to aspire to, like democracy. Paradoxically, these circumstances can pose a threat to democracy itself, as the citizens suspend critical judgment or justify the actions of the authority for the sake of a higher cause. On the other hand, citizens who have an active political orientation based on values are less likely to fall into this trap, and keep a critical eye on the authority, irrespective of who and what the authority represents.